CN105138690B - The method and apparatus for determining keyword - Google Patents
The method and apparatus for determining keyword Download PDFInfo
- Publication number
- CN105138690B CN105138690B CN201510598563.XA CN201510598563A CN105138690B CN 105138690 B CN105138690 B CN 105138690B CN 201510598563 A CN201510598563 A CN 201510598563A CN 105138690 B CN105138690 B CN 105138690B
- Authority
- CN
- China
- Prior art keywords
- commodity
- keyword
- index
- alternative keywords
- key word
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000011156 evaluation Methods 0.000 claims abstract description 23
- 230000002860 competitive effect Effects 0.000 claims description 24
- 238000012545 processing Methods 0.000 claims description 8
- 230000001737 promoting effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000002349 favourable effect Effects 0.000 description 3
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000035699 permeability Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 206010021703 Indifference Diseases 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a kind of method and apparatus of determining keyword.The method of the determining keyword includes:Obtain the raw information of commodity, wherein the raw information includes commodity classification, item property, price situation, sales situation and the evaluation situation of the commodity;Keyword is carried out to the commodity according to the raw information of the commodity and opens up word, obtains the alternative keywords of the commodity;Determine the characteristic index of the alternative keywords;Keyword is selected from the alternative keywords according to the characteristic index, the keyword as the commodity.The method and apparatus of determining keyword provided in an embodiment of the present invention improve the accuracy and timeliness of the keyword determined for commodity.
Description
Technical field
The present embodiments relate to Internet technical field more particularly to a kind of method and apparatus of determining keyword.
Background technology
In recent years, as internet is deeply to every household, shopping at network has become in people's daily life very
A kind of important shopping way.Correspondingly, more and more enterprises also carry out the important of product promotion using internet as them
Means.
The case the most classical of the existing network promotion is Taobao's through train.It is that Taobao makes to measure for buyer
A promotion tool by pay-per-click.This tool can effectively improve on-line shop's exposure rate and conclusion of the business conversion ratio, can
To realize the precision marketing of dotey.When buyer searches for a commodity, as long as containing buyer institute in the keyword of bidding of seller
The word of search, the product of seller are possible in the product list for appearing in Taobao's page.As it can be seen that in the product of Taobao's through train
In the network promotion, a most key ring is exactly to choose suitable keyword for product.Tool is promoted in existing networking products
In, the keyword selection of product is manually completed.The disadvantage done so is, chooses subjectivity when keyword
Greatly, not accurate enough, it is difficult to complete the purpose of precision marketing.
Invention content
In view of the above technical problems, the embodiment of the present invention proposes a kind of method and apparatus of determining keyword, true to improve
The accuracy of fixed keyword.
In a first aspect, an embodiment of the present invention provides a kind of method of determining keyword, the method includes:
Obtain the raw information of commodity, wherein the raw information include the commodity classifications of the commodity, item property,
Price situation, sales situation and evaluation situation;
Keyword is carried out to the commodity according to the raw information of the commodity and opens up word, obtains the alternative key of the commodity
Word;
Determine the characteristic index of the alternative keywords;
Keyword is selected from the alternative keywords according to the characteristic index, the keyword as the commodity.
Second aspect, an embodiment of the present invention provides a kind of device of determining keyword, described device includes:
Data obtaining module, the raw information for obtaining commodity, wherein the raw information includes the quotient of the commodity
Category mesh, item property, price situation, sales situation and evaluation situation;
Word module is opened up, word is opened up for carrying out keyword to the commodity according to the raw information of the commodity, obtains described
The alternative keywords of commodity;
Index determining module, the characteristic index for determining the alternative keywords;
Keyword Selection module, for selecting keyword from the alternative keywords according to the characteristic index, as
The keyword of the commodity.
The raw information that the method and apparatus of determining keyword provided in an embodiment of the present invention pass through acquisition commodity, wherein
The raw information includes commodity classification, item property, price situation, sales situation and the evaluation situation of the commodity, root
Keyword is carried out to the commodity according to the raw information of the commodity and opens up word, the alternative keywords of the commodity is obtained, determines institute
The characteristic index for stating alternative keywords selects keyword, as described according to the characteristic index from the alternative keywords
The keyword of commodity improves the accuracy of determining keyword, realizes the precision marketing to product.
Description of the drawings
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, of the invention other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is the flow chart of the method for the determination keyword that first embodiment of the invention provides;
Fig. 2 be the determination keyword that second embodiment of the invention provides method in open up the flow chart of word;
Fig. 3 be third embodiment of the invention provide determination keyword method in characteristic index determine flow chart;
Fig. 4 is the functional relation schematic diagram between the affiliated rate that third embodiment of the invention provides and matching order;
Fig. 5 is the flow chart of the method for the determination keyword that fourth embodiment of the invention provides;
Fig. 6 be fifth embodiment of the invention provide determination keyword method in Keyword Selection flow chart;
Fig. 7 is the structure chart of the device for the determination keyword that sixth embodiment of the invention provides.
Specific implementation mode
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is only used for explaining the present invention rather than limitation of the invention.It also should be noted that for the ease of
Description, only some but not all contents related to the present invention are shown in the drawings.
First embodiment
Present embodiments provide a kind of technical solution of the method for determining keyword.In the technical scheme, the determination
The method of keyword includes:Obtain commodity raw information, wherein the raw information include the commodity commodity classification,
Item property, price situation, sales situation and evaluation situation;The commodity are closed according to the raw information of the commodity
Keyword opens up word, obtains the alternative keywords of the commodity;Determine the characteristic index of the alternative keywords;Referred to according to the feature
Number selects keyword from the alternative keywords, the keyword as the commodity.
Referring to Fig. 1, the method for the determining keyword includes:11 are operated to operation 14.
Operation 11 in, obtain the raw information of commodity, wherein the raw information include the commodity commodity classification,
Item property, price situation, sales situation and evaluation situation.
One is determined when the keyword for the commodity promoted, the raw information of the commodity should be obtained first.It is described original
Information is to choose basic data based on keyword to commodity.There is no above-mentioned raw information, can not just determine that a word is
No should be the keyword for describing the commodity.
The raw information includes commodity classification, item property, price situation, sales situation and the evaluation of the commodity
Situation.The commodity classification is to confer to an attribute of the commodity, and it can be according to prepared in advance, multi-layer quotient
Product category list and provide.For example, the commodity classification of a casual trousers for men can be " men's clothing-casual pants ".
The item property is determined according to the commodity category information of commodity.For example, if the commodity classification of the commodity
It is clothes, then the item property of the commodity can be the size of the commodity, style etc.;If the commodity class of the commodity
Mesh is PC, then the item property of the commodity can be the model of PC, CPU configurations etc..
The price situation of the commodity refers to the price of a commodity.In the present embodiment, the price of commodity is with commodity
The form of RMB price provides.For with the commodity price of other currency denominations other than RMB, needing other currency
The price of expression is converted into as RMB price.
The sales situation of the commodity refers to the number of packages that a commodity are averagely sold in unit interval.For example, one
Commodity are in one month time, and average commodity daily sell 5, then the sales situation of the commodity is 5/day.
In addition, the sales situation of the commodity further includes the sale number of packages of the commodity, selling time, shelf life, moves back
Goods situation, collection etc..
The evaluation situation of the commodity refers to after commodity are sold, and user is by internet to the evaluation situation of commodity.Specifically
, the evaluation situation of the commodity can show as favorable comment ratio or poor comparation and assessment rate.For example, the evaluation about a commodity
In, 60% is favorable comment, then the favorable comment ratio of the commodity is 60%.
Further, the raw information further includes the commodity title of the commodity, commercial specification, remaining number of packages, shows
The parameters such as amount, click volume, average ranking.
In operation 12, keyword is carried out to the commodity according to the raw information of the commodity and opens up word, obtains the commodity
Alternative keywords.
On the promotion tool platform, the commodity class belonging to each commodity can bind several keywords now.
For example, under commodity classification " men's clothing-casual pants ", the keywords such as " cultivating one's moral character ", " straight tube " can have been bound.It is promoted getting to wait for
Commodity raw information after, the keyword of commodity class now belonging to the commodity is obtained, as promoting the commodity
Keyword.
In addition, after getting the raw information of commodity of seller's upload, promotion tool platform can also obtain
The alternative keywords about the commodity that seller is further arranged.Such operation is primarily to facilitate seller to be pushed away in marketing
Need the commodity promoted that free keyword is set for oneself on wide tool platform, as the keyword for promoting the commodity.
But if the keyword for only binding affiliated commodity class now and the free keyword of seller oneself setting
Keyword when as the popularization commodity, usually because the select permeability of keyword leads to the popularization effect promoted to commodity
Fruit is bad.For example, the keyword of selection is excessively common, the unique features or keyword of the commodity for needing to promote can not be embodied
It is improper to select so that selected keyword can not accurately indicate the characteristics of commodity.In this case, although to commodity into
Go popularization, but promotion effect and bad.Specific manifestation is, after being promoted to the commodity, to the commodity
Pageview and exchange hand are not obviously improved.
In order to overcome the select permeability of above-mentioned keyword, word operation is opened up to the keyword execution of the commodity.Word is opened up in execution
The purpose of operation is the quantity for the keyword for including in the keyword set for expanding the commodity.Specifically, can select with
The primary keys of the commodity have alternative keywords of the word of same or similar semanteme as the commodity, to complete to open up
The operation of word.
Further, word can be searched by obtaining the heat about primary keys that provide of search engine, to obtain and institute
Stating primary keys has the word of same or similar semanteme.For example, search engine is provided about primary keys " straight tube "
It is " straight barrel type " that heat, which searches word, then the heat can be searched to word " straight barrel type " the also alternative keywords as the commodity.
In operation 13, the characteristic index of the alternative keywords is determined.
After opening up word operation, the quantity of the alternative keywords of the commodity, which has, significantly to be increased.But described
In the alternative keywords of commodity, contacting between some alternative keywords and the commodity is simultaneously less close, so buyer exists
The probability scanned for using the alternative keywords when searching for the commodity of same type is very small.It is therefore desirable to from the commodity
Suitable keyword is filtered out in alternative keywords.
In order to filter out suitable keyword from the alternative keywords, it is thus necessary to determine that carry out alternative keywords and screen institute
The index of foundation, that is, the alternative keywords characteristic index.The characteristic index includes:Of the alternative keywords
With index, show index, click at least one of index and competitive index index.It is corresponding, determine the alternative keywords
Characteristic index include the match index of the determining alternative keywords, show index, click in index and competitive index extremely
A few index.
In operation 14, keyword is selected from the alternative keywords according to the characteristic index, as the commodity
Keyword.
The characteristic index of the alternative keywords is determined, has also determined that finger based on carry out alternative keywords screening
Mark.Therefore, keyword can be selected from the alternative keywords according to the characteristic index, to complete the pass of the commodity
The screening of keyword.
The present embodiment closes the commodity according to the raw information of the commodity by the raw information of acquisition commodity
Keyword opens up word, obtains the alternative keywords of the commodity, determines the characteristic index of the alternative keywords, and according to the spy
Sign index selects keyword from the alternative keywords, as the keyword of the commodity, to according to the alternative key
The characteristic index of word selects the keyword more particularly suitable to product promotion, improves the accuracy of determining keyword.
Second embodiment
The present embodiment further provides the original letter according to the commodity based on the above embodiment of the present invention
Breath carries out keyword to the commodity and opens up word, obtains a kind of technical solution of the alternative keywords of the commodity.In the technical side
In case, keyword is carried out to the commodity according to the raw information of the commodity and opens up word, obtains the alternative keywords of the commodity
Including:Obtain the first qualifying key word of the setting quantity of the commodity class of the commodity now;It will be with first qualifying key
Word justice is relevant and retrieves keyword of the frequency higher than predeterminated frequency thresholding as the second qualifying key word;Will with it is described
Second qualifying key word justice is relevant and retrieves keyword of the frequency higher than predeterminated frequency thresholding as third qualifying key
Word;The qualifying key of repetition is removed from the first qualifying key word, the second qualifying key word and third qualifying key word
Word, and using the qualifying key word after duplicate removal processing as the alternative keywords of the commodity.
Referring to Fig. 2, keyword is carried out to the commodity according to the raw information of the commodity and opens up word, obtains the commodity
Alternative keywords include:21 are operated to operation 24.
Operation 21, obtains the first qualifying key word of the setting quantity of the commodity class of the commodity now.
The qualifying key word is the keyword with the commodity classification binding belonging to the commodity.Obtaining the commodity
Raw information after, the commodity classification for obtaining the commodity divides into fixed number purpose the first qualifying key word.Specifically, can root
According to the inquiry frequency of the keyword setting number is selected among the keyword bound with the commodity classification belonging to the commodity
Keyword, as the first qualifying key word.Further, the specific value of the setting number is 100.
Operation 22, will it is relevant with the first qualifying key word justice and retrieve frequency be higher than predeterminated frequency thresholding
Keyword as the second qualifying key word.
The purpose for opening up word operation is based on the first qualifying key word, by further choosing and described the
The same or similar word of semanteme of one qualifying key word is as opening up word as a result, so that the quantity of the keyword of the commodity
Increase.So when open up word operation, should choose it is relevant with the first qualifying key word justice, that is, semantically
It is used as with the same or similar word of the first qualifying key word and opens up word result.
When using a term by search engine retrieving internet web page contents, search engine is being returned about institute
When stating the retrieval result of term, it can return simultaneously and search word with the semantically related heat of the term.Also, it is public that heat, which searches word,
Crowd uses the word being commonly used in process in actual search engine, that is to say, that the heat searches the retrieval frequency of word higher than pre-
If frequency threshold.Since the heat that described search engine returns is searched between word and the term of script in semantically identical or phase
Closely, so searching word by the heat of acquisition the first qualifying key word to obtain the second qualifying key word, to completion pair
The first qualifying key word opens up word.
In order to get the second more qualifying key word, described can be obtained in a manner of word for word increased
The heat of one qualifying key word searches word, and the heat got in this way is searched word as the second qualifying key word.For example,
The first qualifying key word is " autumn and winter money ", then the heat that can obtain " autumn ", " autumn and winter " and " autumn and winter money " respectively searches word, and
The heat of these three words is searched into word and is all used as the second qualifying key word.
Operation 23, will it is relevant with the second qualifying key word justice and retrieve frequency be higher than predeterminated frequency thresholding
Keyword as third qualifying key word.
It is identical as the principle in aforesaid operations 22, search engine can be utilized, by obtaining the second qualifying key word
Heat search word to obtain the third qualifying key word, to complete to open up word operation to the second qualifying key word.
In addition, it is identical as aforesaid operations 22, when obtaining third qualifying key word, word for word increased mode can also be used
It carries out heat and searches word acquisition, and the heat got in this manner is searched into word as the third qualifying key word.
Operation 24 removes weight from the first qualifying key word, the second qualifying key word and third qualifying key word
Multiple qualifying key word, and using the qualifying key word after duplicate removal processing as the alternative keywords of the commodity.
The first qualifying key word, the second qualifying key word and the third qualifying key got by aforesaid operations step
In word, it inevitably will appear the keyword of repetition.In order to ensure the uniqueness for the alternative keywords that the commodity are arranged, to above-mentioned
The first qualifying key word, the second qualifying key word and the third qualifying key word execution duplicate removal processing got, that is, from
The qualifying key word of repetition is removed in the first qualifying key word, the second qualifying key word and third qualifying key word.It is right
It, will be remaining pre- after above-mentioned first qualifying key word, the second qualifying key word and third qualifying key word execute duplicate removal processing
Select keyword as the alternative keywords of the commodity.
The present embodiment, will be with institute by obtaining the first qualifying key word of the setting number of the commodity class of the commodity now
State the first qualifying key word justice it is relevant and retrieve frequency higher than predeterminated frequency thresholding keyword as second pre-selection close
Keyword, by it is relevant with the second qualifying key word justice and retrieve frequency higher than predeterminated frequency thresholding keyword make
For third qualifying key word, and from the first qualifying key word, the second qualifying key word and third qualifying key word
The qualifying key word repeated is removed, and using the qualifying key word after duplicate removal processing as the alternative keywords of the commodity,
Word is opened up to the keyword of the commodity to realize, expands the range of the alternative keywords of the commodity.
3rd embodiment
The present embodiment further provides for determining the alternative keywords based on the above embodiment of the present invention
A kind of technical solution with index.In the technical scheme, determine that the match index of the alternative keywords includes:Described in determination
Individual character accounting rate of the alternative keywords in the commodity title of the commodity;Determine the quotient of the alternative keywords and the commodity
Matching rate between product attribute;Determine the negative match rate between the alternative keywords and the negative attribute of the commodity;It will
The commodity classification of the commodity is matched with the category list of the alternative keywords, is obtained the commodity and is belonged to described alternative
The affiliated rate of the category list item of keyword;And according to the individual character accounting rate, matching rate, negative match rate and affiliated rate
Calculate the match index between the alternative keywords and the commodity.
Referring to Fig. 3, determine that the match index of the alternative keywords includes:31 are operated to operation 35.
In operation 31, individual character accounting rate of the alternative keywords in the commodity title of the commodity is determined.
The individual character accounting rate refers to that each individual character in the alternative keywords determined for the commodity is included in the quotient
Ratio in the commodity title of product.For example, the commodity title of the commodity is " low side plate shoes for women city prevalence playshoes 2014
Autumn in year is trendy ", it is " leisure series " if it is the alternative keywords that the commodity are selected, the alternative keywords are included
Four individual characters in, individual character " stopping " and " spare time " are included in the commodity title, and described in individual character " being " and " row " are not included in
In commodity title.So it is 50% that the individual character of the alternative keywords " leisure series ", which includes rate,.
In operation 32, the matching rate between the alternative keywords and the item property of the commodity is determined.
The specific item property of itself of the commodity.The item property may include the price, color, production of commodity
The information such as quotient, size, discount.The match index of the alternative keywords further includes the quotient of the alternative keywords and the commodity
Matching rate between product attribute.
Matching rate between the item property refers to, by the alternative keywords and the item property of the commodity into
Row text matches.If carrying out successful match when text matches between the alternative keywords and the item property of the commodity,
The matching rate is 100%;If when carrying out text matches between the alternative keywords and the item property of the commodity not
With success, then the matching rate is 0%.
In operation 33, the negative match rate between the alternative keywords and the negative attribute of the commodity is determined.
The negative attribute of the commodity refers to the category that described commodity itself do not have, or have with the commodity itself
The property opposite property parameters of parameter value.For example, the color of a T-shirt be red, then its negative attribute can be green,
Blue, black or grey.For another example, the cartridge types of a wrist-watch are electronics, then the cartridge types pair of this wrist-watch
The negative attribute answered is machinery.
After the matching rate between the alternative keywords and the item property is determined, further determine described standby
Select the negative match rate between keyword and the negative attribute of the commodity.The negative match rate be the alternative keywords not
Matched probability occurs between the negative attribute of the commodity.Specifically, can by by the alternative keywords with it is described
Text matches between the negative attribute of commodity are no between the alternative keywords and the negative attribute of the commodity to determine
Determine matching rate.Further, if the negative attributes match success of the alternative keywords and the commodity, the negative
It is 20% with rate;If the alternative keywords and the negative attributes match of the commodity are failed, the negative match rate
It is 100%.
In operation 34, the commodity classification of the commodity is matched with the category list of the alternative keywords, is obtained
The commodity belong to the affiliated rate of the category list item of the alternative keywords.
Similar with the commodity classification that the commodity have, the alternative keywords also have the category list of itself.It is described
The list of categories of alternative keywords is divided according to level, and the list cell in each level has the title of its own.
The alternative keywords are classified into according to the semanteme of its own at least one specific unit of the category list.Cause
This, each alternative keywords are just provided with its affiliated list cell in the category list.
Unlike the commodity classification of the commodity, the same alternative keywords can belong to the category list simultaneously
In multiple list cells.For example, alternative keywords " cultivating one's moral character " may belong to list cell " men's clothing | casual pants ", can also belong to
In list cell " men's clothing | jeans ".
It, can be by the quotient of the commodity since each unit in the commodity classification also has its corresponding title
Category purpose title is matched with the title of the list of categories belonging to the keyword, belongs to described to obtain the commodity
The affiliated rate of the category list item of alternative keywords.
Further, the affiliated rate is obtained by following formula:
Wherein, M is the affiliated rate of the commodity, and X is that the commodity classification is matched with the alternative keywords classification
Matching order.
Fig. 4 shows affiliated functional relation between rate and matching order.Referring to Fig. 4, the commodity classification with it is described standby
Select the matched matching order of list of categories progress of keyword more forward, that is, the commodity classification and the alternative keywords
The value that affiliated list of categories carries out matched matching order is smaller, then the affiliated rate of the commodity is higher.
Assuming that the title of the commodity classification of the commodity is " men's clothing | casual pants ", and the alternative keywords " cultivating one's moral character " institute
The title of the list of categories of category includes:" men's clothing | casual pants ", " men's clothing | jeans ", " men's clothing | Western-style trousers ", " men's clothing | sweater/
Sweater ", " men's clothing | jacket ", " children's garment/parent-offspring's dress | trousers ".Since the classification of the merchandise classification and the alternative keywords arranges
It is 1 that table, which carries out matched matching order, so the value that the commodity belong to the affiliated rate of category list is possible value
In maximum value.
In operation 35, the alternative pass is calculated according to the individual character accounting rate, matching rate, negative match rate and affiliated rate
Match index between keyword and the commodity.
After the individual character accounting rate, matching rate, negative match rate and the affiliated rate that calculate to beat the commodity, according to above-mentioned
Parameters calculate the match index between the alternative keywords and the commodity.Preferably, the match index is under
Formula and calculate:
Pmate=[ω1×α+(1-ω1)×β]×γ×M×1000
Wherein, PmateFor match index, α is the individual character accounting rate, and β is the matching rate, and γ is the negative match
Rate, M is the affiliated rate, and ω1For the weight coefficient of the individual character accounting rate.The value for the match index being finally calculated
Between 0 to 1000.
Individual character accounting rate of the present embodiment by the determination alternative keywords in the commodity title of the commodity determines
Matching rate between the alternative keywords and the item property of the commodity determines the alternative keywords and the commodity
Negate the negative match rate between attribute, by the progress of the category list of the commodity classification of the commodity and the alternative keywords
Match, obtain the affiliated rate that the commodity belong to the category list item of the alternative keywords, according to the individual character accounting rate, matching
Rate, negative match rate and affiliated rate calculate the match index between the alternative keywords and the commodity, to realize
The calculating of characteristic index between the commodity and its alternative keywords, accurately to select the keyword of the commodity to provide base
Plinth.
Fourth embodiment
The present embodiment further provides one kind of the method for determining keyword based on the above embodiment of the present invention
Technical solution.In the technical scheme, according to the match index from the alternative keywords select keyword before, institute
The method of stating further includes:Determine whether to promote the commodity according to the evaluation situation of the commodity;If it is determined that the quotient
Product are promoted, then trigger to execute described selecting keyword from the alternative keywords according to the match index.
Referring to Fig. 5, the method for the determining keyword includes:51 are operated to operation 56.
Operation 51 in, obtain the raw information of commodity, wherein the raw information include the commodity commodity classification,
Item property, price situation, sales situation and evaluation situation.
In operation 52, keyword is carried out to the commodity according to the raw information of the commodity and opens up word, obtains the commodity
Alternative keywords.
In operation 53, the characteristic index of the alternative keywords is determined.
In operation 54, determine whether to promote the commodity according to the evaluation situation of the commodity.
The range of goods sold on internet is various, and some of which commodity are easy to be favored by consumer, are more suitble to
It is promoted, and other commodity be not easy to for various reasons attract consumer concern, be not to they into
Row is promoted.
Judge that a commodity can be sentenced if appropriate for popularization is carried out by the evaluation situation of the commodity on the internet
It is disconnected.Specifically, if the difference comparation and assessment of a commodity can be determined that the commodity are not appropriate for carrying out more than difference comparation and assessment threshold value
It promotes, and if the difference comparation and assessment of a commodity are less than difference comparation and assessment threshold value, it can be determined that the commodity are appropriate for promoting.It is preferred that
, the difference comparation and assessment threshold value is 10%.
It is judged as being not suitable for popularization according to the evaluation situation of the commodity if businessman wants the commodity promoted originally,
Other one then can be selected from the commodity of the commodity same type carries out examination popularization.Specifically, can according to it is original
The price situation selection of the commodity of commodity same type needs the commodity for trying to promote.
In the commodity for selecting examination to promote according to price situation, a point position can be carried out to the price situation of commodity, then root
The commodity for needing to try to promote are selected according to the result of position is divided.For example, the quantile of highest price in similar commodity is determined as 1,
The price of other same type commodity is normalized using this highest price, obtains the price point of other same type commodity
Digit.
The price for the commodity that can price quantile be in 0.1 to 0.2 is determined as that inexpensively, price quantile being in
The price of 0.2 to 0.4 commodity is determined as relatively cheap, and the price that price quantile is in 0.4 to 0.7 commodity is determined as
Normally, the price that price quantile is in 0.7 to 0.9 commodity is determined as slightly expensive, the quotient by price quantile 0.9 to 1
The price of product is determined as costliness.
After being determined that the price of the commodity divides bit attribute according to the above process, divide in bit attribute from above-mentioned several prices
One price of selection divides bit attribute, to determine the commodity for needing to try to promote.Preferably, it should select price that bit attribute is divided to be to compare
Cheap commodity are as the commodity for needing to try to promote.
In operation 55, however, it is determined that the commodity are promoted, then trigger execute it is described according to the characteristic index from institute
It states and selects keyword in alternative keywords.
If according to the evaluation situation of the commodity, determination the commodity are promoted, then according to the characteristic index from
The keyword of the commodity is selected in the alternative keywords, and the commodity are promoted according to selected keyword.
In operation 56, keyword is selected from the alternative keywords according to the characteristic index, as the commodity
Keyword.
The present embodiment by according to the match index from the alternative keywords select keyword before, according to institute
The evaluation situation for stating commodity determines whether to promote the commodity, and when the commodity are promoted in determination, touches
Hair execution is described to select keyword according to the match index from the alternative keywords, to execute popularization behaviour to commodity
Before work, judge that selected commodity whether suitable for carrying out the network promotion, further ensure that commodity carry out the network promotion
Effect.
5th embodiment
The present embodiment is further provided according to the characteristic index based on the above embodiment of the present invention from described
A kind of technical solution of keyword is selected in alternative keywords.In the technical scheme, according to the characteristic index from described standby
It selects and selects keyword in keyword, including:According to the sales situation of the commodity, determines the match index, shows index, point
Hit the weight of index and competitive index;According to the match index, show index, click index and competitive index and its weight,
It obtains agreeing with index between the alternative keywords and the commodity;And the highest N number of alternative keywords of index will be agreed with
Keyword as the commodity.
Referring to Fig. 6, the keyword is selected from the alternative keywords to include according to the characteristic index:61 are operated to operation
63。
In operation 61, according to the sales situation of the commodity, determine the match index, show index, click index and
The weight of competitive index.
Calculate between the commodity and the alternative keywords when agreeing with index, match index shows index, clicks
Index and competitive index have been assigned different weight coefficients respectively.By the commodity when in order to agree with index described in calculating
Sales situation take into account, according to the sales situation of the commodity adjust the match index, show index, click index with
And the weight coefficient of competitive index respectively.
To the match index, show index, click before index and competitive index carry out weight coefficient adjustment, need
To specify the sales situation of the commodity.Specifically, number of packages can be sold as reference, to other using highest day in all commodity
The day sale number of packages of commodity is normalized, and obtains the sale quantile of other commodity.If the sale quantile of a commodity
Between 0.1 to 0.4, then judge the sales situation of the commodity for indifference;If the sale quantile of a commodity 0.4 to
Between 0.7, then judge the sales situation of the commodity for middle reaches;If the sale quantile of a commodity between 0.7 to 0.9,
Then judge the sales situation of the commodity for fast sale;If the sale quantile of a commodity judges the quotient between 0.9 to 1
The sales situation of product is quick-fried money.
If the sales situation of a commodity is to be judged as quick-fried money, calculate between the commodity and its alternative keywords
The weight coefficient of competitive index should be reduced when agreeing with index, and increases the weight coefficient for showing index.
In operation 62, according to the match index, shows index, click index and competitive index and its weight, obtain institute
It states and agrees with index between alternative keywords and the commodity.
Agree with index between the alternative keywords and the commodity specifically, calculating according to the following formula:
Wherein, IDegree is to agree with index, PmateFor the match index, click is the keyword click volume, imp
For the keyword amount of showing, contend is the competitive index, w2For the weight coefficient of the match index, w3For the pass
The weight coefficient of keyword click volume, that is, click the weight coefficient of index, w4For the weight coefficient of the keyword amount of showing,
Namely show the weight coefficient of index, w5For the weight coefficient of the competitive index.Further, a is the match index
PmateThe minimum standard that should reach, b are the minimum standard of the keyword amount of the showing imp, and c is the keyword click volume
Minimum standard inverse.
The weight coefficient of the weight coefficient of the match index, the weight coefficient of keyword click volume, the keyword amount of showing
And the weight coefficient of competitive index can adaptively be adjusted according to the sales situation of the commodity.
In operation 63, the highest N number of alternative keywords of index will be agreed with as the keyword of the commodity.
Be calculated between the alternative keywords and the commodity agree with index after, according to it is described agree with index will
The alternative keywords carry out descending arrangement, and will wherein agree with the highest N number of alternative keywords of index as the commodity
Keyword.
The present embodiment is determined the match index, is showed index, clicks index by the sales situation according to the commodity
With the weight of competitive index, according to the match index, shows index, click index and competitive index and its weight, obtain institute
The index that agrees between alternative keywords and the commodity is stated, and the highest N number of alternative keywords of index will be agreed with as institute
The keyword for stating commodity improves selected to realize the selection according to the keyword for agreeing with commodity described in exponent pair
The accuracy of keyword.
Sixth embodiment
The present embodiment provides a kind of technical solutions for the device for determining keyword.Referring to Fig. 7, the dress of the determining keyword
Set including:Data obtaining module 71 opens up word module 72, index determining module 73 and Keyword Selection module 86.
Described information acquisition module 71 is used to obtain the raw information of commodity, wherein the raw information includes the quotient
Commodity classification, item property, price situation, sales situation and the evaluation situation of product.
The word module 72 of opening up opens up word for carrying out keyword to the commodity according to the raw information of the commodity, obtains
The alternative keywords of the commodity.
The index determining module 73 is used to determine the characteristic index of the alternative keywords.
The Keyword Selection module 76 is used to select keyword from the alternative keywords according to the characteristic index,
Keyword as the commodity.
Further, the word module of opening up includes:Keyword pre-selection submodule, first open up lexon module, second open up lexon
Module and keyword determination sub-module.
The keyword pre-selection submodule is used to obtain the first pre-selection of the setting quantity of the commodity class of the commodity now
Keyword.
Described first opens up lexon module for will be relevant with the first qualifying key word justice and retrieving frequency height
In predeterminated frequency thresholding keyword as the second qualifying key word.
Described second opens up lexon module for will be relevant with the second qualifying key word justice and retrieving frequency height
In predeterminated frequency thresholding keyword as third qualifying key word.
The keyword determination sub-module is used for pre- from the first qualifying key word, the second qualifying key word and third
The qualifying key word that repetition is removed in keyword is selected, and using the qualifying key word after duplicate removal processing as the standby of the commodity
Select keyword.
Further, the index determining module 73 includes:Characteristic index determination sub-module.
The characteristic index determination sub-module is used to determine the match index of the alternative keywords, shows index, clicks
At least one of index and competitive index index.
Further, it is specific to use when the characteristic index determination sub-module determines the match index of the alternative keywords
In:
Determine individual character accounting rate of the alternative keywords in the commodity title of the commodity;
Determine the matching rate between the alternative keywords and the item property of the commodity;
Determine the negative match rate between the alternative keywords and the negative attribute of the commodity;
The commodity classification of the commodity is matched with the category list of the alternative keywords, obtains the commodity category
In the affiliated rate of the category list item of the alternative keywords;And
The alternative keywords and institute are calculated according to the individual character accounting rate, matching rate, negative match rate and affiliated rate
State the match index between commodity.
Further, the match index is obtained by following formula:
Pmate=[ω1×α+(1-ω1)×β]×γ×M×1000
Wherein, PmateFor match index, α is the individual character accounting rate, and β is the matching rate, and γ is the negative match
Rate, M are the affiliated rate.
Further, the affiliated rate is obtained by following formula:
Wherein, M is the affiliated rate of the commodity, and X is that the commodity classification is matched with the alternative keywords classification
Matching order.
Further, the device of the determining keyword further includes:It promotes determination module 74 and promotes trigger module 75.
The popularization determination module 74 from the alternative keywords according to the characteristic index for selecting keyword
Before, determine whether to promote the commodity according to the evaluation situation of the commodity.
The popularization trigger module 75 from the alternative keywords according to the characteristic index for selecting keyword
Before, however, it is determined that the commodity are promoted, triggering execute it is described according to the characteristic index from the alternative keywords
Select keyword.
Further, the popularization determination module 74 includes:Certainly decision sub-module and negative decision sub-module.
Certainly the decision sub-module is used for when the difference comparation and assessment of the commodity are greater than or equal to 10%, it is determined that described
Commodity are not suitable for promoting.
The negative decision sub-module is used for when the difference comparation and assessment of the commodity are less than 10%, it is determined that the commodity are suitable
It closes and promotes.
Further, the Keyword Selection module 76 includes:Weight determination sub-module, agree with index determination sub-module with
And keyword determination sub-module.
The weight determination sub-module is used for the sales situation according to the commodity, determines the match index, shows finger
Number, the weight for clicking index and competitive index.
It is described agree with index determination sub-module for according to the match index, show index, click index and competition and refer to
Number and its weight, obtain agreeing with index between the alternative keywords and the commodity.
The keyword determination sub-module will be for that will agree with pass of the highest N number of alternative keywords of index as the commodity
Keyword.
Further, the index that agrees with is obtained by following formula:
Wherein, IDegree is to agree with index, PmateFor the match index, click is the keyword click volume, imp
For the keyword amount of showing, contend is the competitive index, w2For the weight coefficient of the match index, w3For the pass
The weight coefficient of keyword click volume, w4For the weight coefficient of the keyword amount of showing, w5For the weight system of the competitive index
Number, the minimum that a should reach for matching degree, b are the minimum for the amount of showing, and c is the inverse of clicking rate minimum.
Will be appreciated by those skilled in the art that each module of the above invention or each step can use general meter
Device is calculated to realize, they can be concentrated on single computing device, or be distributed in network constituted by multiple computing devices
On, optionally, they can be realized with the program code that computer installation can perform, so as to be stored in storage
It is performed by computing device in device, either they are fabricated to each integrated circuit modules or will be more in them
A module or step are fabricated to single integrated circuit module to realize.In this way, the present invention is not limited to any specific hardware and
The combination of software.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with
The difference of other embodiment, just to refer each other for the same or analogous part between each embodiment.
The foregoing is merely the preferred embodiment of the present invention, are not intended to restrict the invention, for those skilled in the art
For, the present invention can have various modifications and changes.It is all within spirit and principles of the present invention made by any modification, equivalent
Replace, improve etc., it should all be included in the protection scope of the present invention.
Claims (9)
1. a kind of method of determining keyword, which is characterized in that including:
Obtain the raw information of commodity, wherein the raw information includes the commodity classifications of the commodity, item property, price
Situation, sales situation and evaluation situation;
Keyword is carried out to the commodity according to the raw information of the commodity and opens up word, obtains the alternative keywords of the commodity;
Determine the characteristic index of the alternative keywords;
Keyword is selected from the alternative keywords according to the characteristic index, the keyword as the commodity;
Wherein, keyword is carried out to the commodity according to the raw information of the commodity and opens up word, obtain the alternative pass of the commodity
Keyword, including:
Obtain the first qualifying key word of the setting quantity of the commodity class of the commodity now;
Using it is relevant with the first qualifying key word justice and retrieve frequency higher than predeterminated frequency thresholding keyword as
Second qualifying key word;
Using it is relevant with the second qualifying key word justice and retrieve frequency higher than predeterminated frequency thresholding keyword as
Third qualifying key word;
The qualifying key of repetition is removed from the first qualifying key word, the second qualifying key word and third qualifying key word
Word, and using the qualifying key word after duplicate removal processing as the alternative keywords of the commodity.
2. according to the method described in claim 1, it is characterized in that, determine the characteristic index of the alternative keywords, including:
It determines the match index of the alternative keywords, shows index, clicks at least one of index and competitive index index.
3. according to the method described in claim 2, it is characterized in that, determine the match index of the alternative keywords, including:
Determine individual character accounting rate of the alternative keywords in the commodity title of the commodity;
Determine the matching rate between the alternative keywords and the item property of the commodity;
Determine the negative match rate between the alternative keywords and the negative attribute of the commodity;
The commodity classification of the commodity is matched with the category list of the alternative keywords, the commodity is obtained and belongs to institute
State the affiliated rate of the category list item of alternative keywords;And
The alternative keywords and the quotient are calculated according to the individual character accounting rate, matching rate, negative match rate and affiliated rate
Match index between product.
4. according to the method described in claim 3, it is characterized in that, the match index is obtained by following formula:
Pmate=[ω1×α+(1-ω1)×β]×γ×M×1000
Wherein, PmateFor match index, α is the individual character accounting rate, and β is the matching rate, and γ is the negative match rate, and M is
Rate belonging to described, ω1For the weight coefficient of the individual character accounting rate.
5. according to the method described in claim 3, it is characterized in that, rate belonging to described is obtained by following formula:
Wherein, M is the affiliated rate of the commodity, and X is that the commodity classification carries out matched with the alternative keywords classification
With order.
6. according to the method described in claim 1, it is characterized in that, according to the characteristic index from the alternative keywords
Before selecting keyword, the method further includes:
Determine whether to promote the commodity according to the evaluation situation of the commodity;
If it is determined that promoted to the commodity, then trigger execute it is described according to the characteristic index from the alternative keywords
Select keyword.
7. according to the method described in claim 2, it is characterized in that, being selected from the alternative keywords according to the characteristic index
Keyword is selected, including:
According to the sales situation of the commodity, the power for determining the match index, showing index, clicking index and competitive index
Weight;
According to the match index, show index, click index and competitive index and its weight, obtain the alternative keywords with
Agree with index between the commodity;And
The highest N number of alternative keywords of index will be agreed with as the keyword of the commodity.
8. the method according to the description of claim 7 is characterized in that the index that agrees with is obtained by following formula:
Wherein, IDegree is to agree with index, PmateFor the match index, click is the keyword click volume, and imp is institute
The keyword amount of showing is stated, contend is the competitive index, w2For the weight coefficient of the match index, w3For the keyword
The weight coefficient of click volume, w4For the weight coefficient of the keyword amount of showing, w5For the weight coefficient of the competitive index, a is
The minimum that matching degree should reach, b are the minimum for the amount of showing, and c is the inverse of clicking rate minimum.
9. a kind of device of determining keyword, which is characterized in that including:
Data obtaining module, the raw information for obtaining commodity, wherein the raw information includes the commodity class of the commodity
Mesh, item property, price situation, sales situation and evaluation situation;
Word module is opened up, word is opened up for carrying out keyword to the commodity according to the raw information of the commodity, obtains the commodity
Alternative keywords;
Index determining module, the characteristic index for determining the alternative keywords;
Keyword Selection module, for selecting keyword from the alternative keywords according to the characteristic index, as described
The keyword of commodity;
Wherein, the word module of opening up includes:Keyword pre-selection submodule, first open up lexon module, second open up lexon module and
Keyword determination sub-module;
The keyword pre-selection submodule is used to obtain the first qualifying key of the setting quantity of the commodity class of the commodity now
Word;
Described first opens up lexon module for will be relevant with the first qualifying key word justice and retrieving frequency higher than pre-
If the keyword of frequency threshold is as the second qualifying key word;
Described second opens up lexon module for will be relevant with the second qualifying key word justice and retrieving frequency higher than pre-
If the keyword of frequency threshold is as third qualifying key word;
The keyword determination sub-module is used to preselect from the first qualifying key word, the second qualifying key word and third and close
The qualifying key word of repetition is removed in keyword, and using the qualifying key word after duplicate removal processing as the alternative pass of the commodity
Keyword.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510598563.XA CN105138690B (en) | 2015-09-18 | 2015-09-18 | The method and apparatus for determining keyword |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510598563.XA CN105138690B (en) | 2015-09-18 | 2015-09-18 | The method and apparatus for determining keyword |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105138690A CN105138690A (en) | 2015-12-09 |
CN105138690B true CN105138690B (en) | 2018-11-06 |
Family
ID=54724037
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510598563.XA Expired - Fee Related CN105138690B (en) | 2015-09-18 | 2015-09-18 | The method and apparatus for determining keyword |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105138690B (en) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106997350B (en) * | 2016-01-22 | 2020-11-17 | 创新先进技术有限公司 | Data processing method and device |
CN107608983B (en) * | 2016-07-11 | 2021-03-30 | 阿里巴巴集团控股有限公司 | Title information optimization method, device, equipment and system |
CN107590678A (en) * | 2017-08-07 | 2018-01-16 | 北京京东尚科信息技术有限公司 | Method of Commodity Recommendation and system |
KR101814005B1 (en) * | 2017-08-21 | 2018-01-02 | 인천대학교 산학협력단 | Apparatus and method for automatically extracting product keyword information according to web page analysis based artificial intelligence |
CN110717104B (en) * | 2019-10-11 | 2022-05-20 | 广州市丰申网络科技有限公司 | Keyword advertisement putting automatic negative keyword method and device |
CN111767738A (en) * | 2020-03-30 | 2020-10-13 | 北京沃东天骏信息技术有限公司 | Label checking method, device, equipment and storage medium |
CN113837820A (en) * | 2020-06-23 | 2021-12-24 | 阿里巴巴集团控股有限公司 | Data processing method, device and equipment |
CN112115704B (en) * | 2020-09-11 | 2024-06-04 | 深圳市菲尔兹数据分析师事务所有限公司 | Product query method and device based on e-commerce platform and computer equipment |
CN113505281B (en) * | 2021-06-15 | 2022-11-08 | 北京三快在线科技有限公司 | Search suggestion word generation method and device, electronic equipment and readable storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102446180A (en) * | 2010-10-09 | 2012-05-09 | 腾讯科技(深圳)有限公司 | Commodity searching method and device adopting same |
CN103729359A (en) * | 2012-10-12 | 2014-04-16 | 阿里巴巴集团控股有限公司 | Method and system for recommending search terms |
CN103914492A (en) * | 2013-01-09 | 2014-07-09 | 阿里巴巴集团控股有限公司 | Method for query term fusion, method for commodity information publish and method and system for searching |
CN104408648A (en) * | 2014-12-09 | 2015-03-11 | 北京博雅立方科技有限公司 | Method and device for choosing items |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8050971B2 (en) * | 2005-10-27 | 2011-11-01 | Nhn Business Platform Corporation | Method and system for providing commodity information in shopping commodity searching service |
-
2015
- 2015-09-18 CN CN201510598563.XA patent/CN105138690B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102446180A (en) * | 2010-10-09 | 2012-05-09 | 腾讯科技(深圳)有限公司 | Commodity searching method and device adopting same |
CN103729359A (en) * | 2012-10-12 | 2014-04-16 | 阿里巴巴集团控股有限公司 | Method and system for recommending search terms |
CN103914492A (en) * | 2013-01-09 | 2014-07-09 | 阿里巴巴集团控股有限公司 | Method for query term fusion, method for commodity information publish and method and system for searching |
CN104408648A (en) * | 2014-12-09 | 2015-03-11 | 北京博雅立方科技有限公司 | Method and device for choosing items |
Also Published As
Publication number | Publication date |
---|---|
CN105138690A (en) | 2015-12-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105138690B (en) | The method and apparatus for determining keyword | |
US11062372B2 (en) | Method for relevancy ranking of products in online shopping | |
CN102841946B (en) | Commodity data retrieval ordering and Method of Commodity Recommendation and system | |
TWI512653B (en) | Information providing method and apparatus, method and apparatus for determining the degree of comprehensive relevance | |
CN108550052A (en) | Brush list detection method and system based on user behavior data feature | |
US9489688B2 (en) | Method and system for recommending search phrases | |
US8510298B2 (en) | Method for relevancy ranking of products in online shopping | |
CN102279851B (en) | Intelligent navigation method, device and system | |
CN109816482B (en) | Knowledge graph construction method, device and equipment of e-commerce platform and storage medium | |
CN104504159B (en) | Application of the positive and negative sequence pattern of multiple supports in customers buying behavior analysis | |
CN107862566A (en) | A kind of Method of Commodity Recommendation and system | |
US20130006713A1 (en) | Method for aggregating pricing information and assigning a fair market value to goods sold in a peer-to-peer e-commerce transaction | |
CN107122980A (en) | The method and apparatus for recognizing the affiliated classification of commodity | |
CN108665333A (en) | Method of Commodity Recommendation, device, electronic equipment and storage medium | |
CN104679771A (en) | Individual data searching method and device | |
CN103778553A (en) | Commodity attribute recommendation method and commodity attribute recommendation system | |
CN105931082B (en) | Commodity category keyword extraction method and device | |
CN107256513A (en) | Method and device is recommended in a kind of collocation of object | |
US10832305B1 (en) | System and method for image processing and searching for classification in a product database | |
CN110288362A (en) | Brush single prediction technique, device and electronic equipment | |
CN106326318A (en) | Search method and device | |
CN107766229B (en) | Method for evaluating correctness of commodity search system by using metamorphic test | |
CN114463089A (en) | Method and device for improving shop visit degree | |
CN106202517A (en) | A kind of online commodity based on big data sort method on webpage | |
CN110968670B (en) | Method, device, equipment and storage medium for acquiring attributes of popular commodities |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20181106 |